People obessed with Grand Theft Auto have gone on killing sprees.
Who are these people and their supposed killing sprees? Do you have evidence to backup this statement or did this just fly out of your ass? Can you prove any cause/effect with GTA and killing sprees? Remember that the various versions of GTA have sold TENS OF MILLIONS of copies total. Even if there were a very low likelihood of GTA directly inducing killing sprees (say 1 in 100,000), you would have at least 100 examples to cite with direct evidence.
It's great that there is competition out there now for iPods but 95% of people who already own iPods just use iTunes to rip to AAC for their libraries and will never buy anything else that isn't compatible with all the hours they've invested in building their music libraries.
There is subjectivity in the world of science. Emotions do come into play. The latest theories are too often presented as fact.
I would prefer the word Hypothesis here to Theory. In Science, the word Theory is used to describe something that is generally accepted as True beyond reasonable doubt where Hypothesis is used to describe a conjecture which has yet to be proven. In Layman's terms, the word Theory occupies the same vernacular niche as the scientists' term Theory. This confusion is often what leads ID designers to say "Evolution is only a theory". But then again, Gravity is *only* a Theory in Scientific terms as well.
They were able to get 78.9 FPS with Prey but only with 4-way full AFR (Alternate Frame Rendering). There is a big issue with serious gamers and this configuration.
On 4-way AFR the driver builds a display list and sends it to a GPU that isn't busy. It is possible to have all four GPUs busy (rendering frames) while the current frame is being displayed and a new display list is being generated by the CPU. This means what you see with 4-way full AFR can be up to 5 frames later than what is going on in the game engine. At 78.9 FPS, this can translate into nearly 64 ms of latency which is enough to get you killed if you're a serious gamer.
Serious gamers with Quad-SLI are going to want to use SFR (Split Frame Rendering) which cuts latency quite a bit but takes a performance hit to the FPS. There are definite inefficiencies to 4-way SFR with having all four cards render portions of the same scene vs 4-way AFR. You generate 4 times as much display list info (GPU fifo data) and you have to replicate more data and uploads (if textures don't fit into the 512MB memory) across the GPUs.
I'm not sure if the Quad-SLI supports an AFR/SFR hybrid where you can have 2X2 (2 GPUs working on SFR each in AFR queue) - this might balance the performance vs latency issue better.
If you have your XP home CD, you should be able to install it.
Bzzzt... most computers shipping XP home don't come with the CD - only a restore to original image program taking up a hidden partition on the HD. Think Dell, Gateway, HP etc. These users don't have backup and can't install it.:(
Windows Backup is not installed on the Home Edition of XP that ships by default on 90% of users' new computers. Plus if you order from Dell or somewhere nice like that, they don't give you an installation CD to install the program (which on the MS website they tell you to hunt down Ntbackup.msi on the CD and run it to install). Plus it does back-up's by folder (which means knowing where the obscure stuff is if it's not in your My documents). Plus it doesn't do selective registry backup and restore... etc.
In otherwords, it's good for saving your "My Documents" folder after you've bothered to install it.
How about Time Machine? This is a very user friendly backup concept. Imagine a normal user performing backups on Windows. How hard is it to save off apps that rewrite 1,000 entries in your registry? Heck your average user has no idea where to find and backup his e-mail:
If you're using Outlook, they're in: C:\Documents and Settings\username\Local Settings\Application Data\Microsoft\Outlook
If you're using Outlook Express, they're in: C:\Documents and Settings\username\Local Settings\Application Data\Identities\{45E80F8D-E0D8-48D0-B459-408C2E2F8 7C8}\Microsoft\Outlook Express
That is unless you've modified (either on purpose or accidentally) some obscure registry keys that redirect these files.
Grammar/spelling nazi mode on On a side note, you don't need to use apostrophes with acronyms when showing the plural form and the market for games on Macs is nonexistent, not inexistant. Carry on. Grammar/spelling nazi mode off.
Mr. Grammar Nazi, you are wrong on both counts. My english is far from perfect but when you wish to correct someone's grammar, it helps if you are writing proper English yourself. Both the items you attempt to correct in my post are proper English (with a capital 'E').
I direct you to the following dictionary entry from Webster's Revised Unabridged Dictionary (1913) : http://dict.die.net/inexistent/
Inexistent \In`ex*ist"ent\, a. [Pref. in- in + existent: cf. F.
inexistant.]
Not having being; not existing.
And it is perfectly OK to use an apostrophe with an acronym to separate the plural form. As a matter of fact, it is MORE proper to do so. However, some people have an aversion to the proper usage of apostrophes with acronyms due to the fact that it violates the rule that apostrophes are not used before an S indicating a plural. Alas, proper English has many exceptions to the rule and here is one of them, esoteric as it may be. For example, why doesn't "i before e except after c" apply to weigh or neighbor? Anyhow, I direct you to the following site at Washington State University: http://www.wsu.edu/~brians/errors/acronyms.html
God I hate being right all the time. (Just Kidding!)
I look forward to your apology and if I ever should meet you in the real world, I will expect you to buy me a beer. In the meantime, I am glad to have enlightened a Grammar Nazi's understanding of English so that he may refrain from grammar-bashing poor fellows writing proper English in his future.
I like programming video games and I like working on games for PC's and Consoles which is what I'm paid to do. I also said I think Macs are nice sexy stable platforms. Heck I even aggressively recycle and drive a fuel-efficient car too which you seem to think makes you special. However, I honestly believe that Macs are not a big enough market for $20 million dollar games to be developed specifically for them and for those games to be profitable. Game development is not the same as solving the world's hunger problems. If Mac owners are hungry for games but their market isn't profitable, it's not a crime against humanity to let them "starve" for new games. We develop games to make money, not as charity.
I have no personal desire to throw away my career or my company's money on developing a project projected to lose millions of dollars regardless of how sexy it is. It's hard enough to make money on PC's and Consoles as it is without targeting a niche market without a distribution channel. And equating my reluctance to push for more game development on an inexistant market to the "worst atrocities, of all kinds... in history" just shows how out of touch with reality you are. I'm perfectly happy where I am and I see no justification to risk my company's financial health with an issue that no one I work with believes in including myself. I want my games to be played by millions of people everywhere (the last one I worked on had 2+ million copies sold).
There's no reason to throw away a happy life, a career, and millions of dollars on some eco-freak hippie whiners like you who think I should "grow a pair" and service them.
Macs have, depending on who you ask, anything up to a 15% market share.
Well, the people I would ask actually write industry reports for video game distribution. I'm assuming the people you ask are other Mac owners living in some fantasy world. Mac video game sales make up significantly less than 1% of the total sales of console and PC games. If they made up 15% of the total video game sales market, they would be more than PC games which make up only 12-14% of the video game market. Most game sales are on consoles (including handhelds) and then on PCs. The Macs are way out on the long end of the tail next to Linux for game sales. If Macs made as big a market share as you'd like to believe, it'd be easier to find games for the Mac than for the GameCube.
I want us to live a more sustainable lifestyle and I practice that, including recycling far more than most people do and driving a small fuel-efficient car and minimizing my driving and living close to where I work and working for a responsible company (a university). I actively encourage those I know to do the same.
There's a pretty big difference between you driving a fuel efficient car and solving the entire country's oil addiction. But dude, "grow a pair" and fix the country's energy problems by yourself. And while you're at it, "grow a pair" and bring peace to the middle east. I know you want to do it but the only thing holding you back is lack of balls.
Now there's a valid solution to all the real world problems with the Mac video game development issues. Why didn't I think of that before?
Attitudes of Mac owners like this (I'll assume you own one by your comment) do not endear you to game developers with much larger gaming markets (consoles ~ $8 billion, PCs ~ $1 billion, Macs ~ $20 million, Linux ~ $1 million) to focus on. When consoles and PC's have 99% of game sales, there's not much point to supporting other platforms -- at least here in the real world that makes sense.
When you're writing stuff for PC and XBOX (or XBOX 360), you use DirectX not OpenGL. PS2 usually uses a custom renderer or a package like Renderware. Only PS3 uses a graphics API similar to OpenGL and most people are going to bypass that for the low-level graphics API for performance reasons. Game Development IDES like Visual Studio (XBOX/XBXO360), SN (PS2/PS3), and CodeWarrior (PS2/PSP/etc) have licensing controls built-in that *ONLY* allow them to run on specific Windows PCs (either by MAC # or with floating licenses off an attached network server). Additional debugging tools (PIX, CATS, WinPACon, etc.) only run on Windows. At one time, Sony actually made a lot of their PS2 tools run on Linux but since developers had to use Windows for everything else, this effort hindered the tool usage until they made cygwin ports for windows. Finally, no one in the industry has made any serious effort to get the game-programming specific development tools we use to run on Macs.
Macs are sexy and nice, they're just not practical for my job. Sorry...
You ever think that you're in a position to, y'know, change that?
To be honest, in my current position as a game programmer for a major company, I have almost no control over the actual platforms we target for game design for current gen or next gen. These decisions are made much higher by management and marketing based on potential market and profitability. When it costs $10-20 million to make a game, no one wants to target a small unproven market. The best they'll risk is low-cost ports after the main game is written. In other words, incremental work that may increase the ROI from the already complete game. Find me a single publisher who will fund a 20-30 person team on a Mac game for two or more years because a programmer told them to do so and I'll eat my words. That's the effort going into high-end PC and console games right now and the sad facts of the economics.
I desire my PC to be pure of spyware, security flaws and unstability.
Somehow Macs do this and manage to be sexy at the same time. That said, I have a PC at work and home. I'd buy a Mac for home except that Macs are sadly lacking in the one field in which I work (video game development).
4) At this point, you have no debt, and you have reached your "margin of safety" amount. Once you have built up an additional $3k to $5k on top of your margin of safety, open up a discount brokerage account (e.g., E-Trade).
Don't use E-Trade. They have so many fees it's impossible to count them all. They charge $19.95 per trade plus some other fees (ends up about $23-25 a trade for a casual investor), they charge to get money out, and they charge you about $60 to CLOSE your account!!! Try Scottrade (www.scottrade.com). Scottrade has lots of local offices (three of them within 5 miles of my house in the NW burbs of Chicago) and they only charge $7 a trade. Plus Scottrade has better margin rates and will pay higher interest (although still low) on any cash sitting in your account.
FWIW, I'm a PS3 developer and the preferred method (i.e. generates the fastest code for the general case) for programming SPU's is to use C with intrinsics. C mind you... not C++. Heavy use of esoteric C++ features on the SPU's just generates some of the worst performing binary code you might ever see. Plus, there are lots of compiler specific (GCC) tricks though. A great place to see some of the performance tricks on CELL is on PS3 developer Mike Acton's site http://www.cellperformance.com/
I prefer to use general optimizations first over compiler specific ones but if top performance is required (and in video games it often is) you have to occasionally make sacrifices of portability and code readability for speed in low level routines.
US Secretary of Education Rod Paige, a member of the Bush Administration, called the National Education Association, the nation's largest teachers union, a terrorist organization. AFAIK, they've never committed hostage-taking, bombings, death threats, or other terrorist activity. Their biggest crime is using strike tactics for collective bargaining and closing schools when to trying to get better pay and working conditions for our nations Teachers.
Around 300 AD, Christian riots and violence were widespread throughout parts of the Roman Empire to a degree that was disrupting society on a regular basis. Think Paris-car burnings combined with Middle East Anti-American protests on a large scale. Now imagine that happening in the US and tell me what the Bush Administration would call it.
So what do you call the Inquisition and the Crusades? Christianity hardly spread throughout the world by peace. Even in Ancient Rome 2000 years ago, the early Christians caused enough destruction and disturbances that the Bush administration would call them Terrorists today.
The Bible is a historic document in the same way that the Iliad is a historic document. Both are collections of myths and fables that are roughly based on actually occurences in history. Both have supernatural events that were not likely but make for a better story. Science lets us determine which parts are likely to be true (i.e. history) and which parts are likely to be nothing more than myth.
Yes, the likelihood of a group of Jewish fisherman making up a story about a Messiah figure who claimed to be God (blasphemy) and then turning the entire Roman empire upside down in the matter of a few decades is highly unlikely. It is even more unlikely that they would all suffer torture and death to protect a story that is not true. And yet, that is exactly what happened. If anything, this is a strong indication that their story was real. Would you die for something you know to be false?
By your logic, a prophet in the middle east who turned the entire regious upside down, resulting in the rapid conversion of the entire area to the same belief must be correct. ESPECIALLY since he has thousands of men and women lining up to die for his beliefs on a daily basis and receive martyrdom for their cause.
Yes... there you have it. Following your logic, both Christianity and Islam are true. And since Muhammed came afterward Jesus and plenty more people are willing to die for Muhammed, Islam must be "more true" than Christianity.
Do you see the flaws in your logic now or are you converting to Islam?
When people require absolute faith regardless of the overwhelming contrary evidence, they have already sacrificed enough of their own identity and ability to reason that sacrificing their lives is merely the next step of losing themselves to their beliefs. Welcome to the Church of Jim Jones, you'll enjoy the Kool-Aid.
"[...]a few people might complain about losing San Francisco and New York to rising water levels."
As a species though, even as a culture, the long term (100+ years perspective) impact would be minimal. We'd rebuild inland and carry on.
The problem with non-linear systems like the climate is that it's possible that we'd get something like a "Venus-effect" eventually and that might be a bit more threatening than displacing a few hundred million people with flooding.
The fast C code that's executing has already been written.. some of it is in the python interpreter, some it is in the ksh and php interpreters, most of it is in the db2 interpreter.
Not *ALL* the "fast C" code has been written. There's lots of code out there that needs to run faster... some written and some yet to be written. You can use python to string together all the calls to already-written-C-libraries you want but when you have an inner loop of a tight algorithm that isn't in a library fo you, you're going to have to break down and code it in C (or JAVA running through a JIT-native compiler, etc).
It's not broken - it's that way by design and there's nothing to "fix" in order to make a good game on the PS3. There's nothing to "fix" because even if you need to share large buffers between the SPU's and RSX, there's a simple and quite fast copyback mechanism.
The SPU single-read from RSX memory is only meant to be used rarely for occasional one-time reads. What Sony did in alerting developers to the faster method of dealing with larger shared buffers is the right thing to do. Not to mentionat that since a fast read out of video memory is not a feature that other consoles have or even most PC cards support, any algorithms making extensive use of this feature are probably being written from scratch anyway and can very easily incoporate the fast RSX copyback. This is analogous to DirectX on a PC, where when you "lock" a surface for reading, the driver will often do the equivalent copyback from video memory to system memory without you even knowing about it! On the PC, it may even de-swizzle or change formats during the copyback.
Pay attention. The article says that SONY is telling the developers to avoid using local memmory at all - that means, it won't be fixed in the retail version.
There are three types of directly addressable memory on the PS3 and there are three types of processors.
PROCESSOR == MEMORY
PPU == main memory
SPU == (SPU) local memory
RSX == (RSX) local memory (AKA video memory)
What Sony is telling developers NOT to do is to READ from (RSX) local memory (AKA video memory) using the SPU. This would be much more clear if the article mentioned this was local memory on the RSX not the SPU. The PS3 has a very fast write port from SPU to RSX. There is no fast read port from RSX to SPU, and for the most part it isn't necessary.
Most of the time, developers will be using the SPU to generate large data sets (dynamic geometry, etc) to be fed to the RSX similar to how the VU1 fed data to the GS on PS2. For the rare times that the SPU will need to access RSX, the main memory writeback mechanism will suffice for large data accesses (i.e. post process on an entire bitmap) and the slow path can be used for very occasional, single small reads (i.e. check the color of just one pixel). On the PS2, there isn't even a slow path for the reads from video memory to VU1. You'd have to set up a copy-back from GS-memory to main memory using the EE and the write it back into VU1 memory (again using the EE or setting up a DMA with the EE) with lots of waits or interrupt driven synchronizations.
FWIW, I don't believe the XBOX360 has a fast ability for the PPC's to read from their local video memory (embedded on the ATI GPU) either.
Slashdot Mirror
People obessed with Grand Theft Auto have gone on killing sprees.
Who are these people and their supposed killing sprees? Do you have evidence to backup this statement or did this just fly out of your ass? Can you prove any cause/effect with GTA and killing sprees? Remember that the various versions of GTA have sold TENS OF MILLIONS of copies total. Even if there were a very low likelihood of GTA directly inducing killing sprees (say 1 in 100,000), you would have at least 100 examples to cite with direct evidence.
It's great that there is competition out there now for iPods but 95% of people who already own iPods just use iTunes to rip to AAC for their libraries and will never buy anything else that isn't compatible with all the hours they've invested in building their music libraries.
I guess you're a Neal Stephenson fan and want to work for the gov't?
There is subjectivity in the world of science. Emotions do come into play. The latest theories are too often presented as fact.
I would prefer the word Hypothesis here to Theory. In Science, the word Theory is used to describe something that is generally accepted as True beyond reasonable doubt where Hypothesis is used to describe a conjecture which has yet to be proven. In Layman's terms, the word Theory occupies the same vernacular niche as the scientists' term Theory. This confusion is often what leads ID designers to say "Evolution is only a theory". But then again, Gravity is *only* a Theory in Scientific terms as well.
They were able to get 78.9 FPS with Prey but only with 4-way full AFR (Alternate Frame Rendering). There is a big issue with serious gamers and this configuration.
On 4-way AFR the driver builds a display list and sends it to a GPU that isn't busy. It is possible to have all four GPUs busy (rendering frames) while the current frame is being displayed and a new display list is being generated by the CPU. This means what you see with 4-way full AFR can be up to 5 frames later than what is going on in the game engine. At 78.9 FPS, this can translate into nearly 64 ms of latency which is enough to get you killed if you're a serious gamer.
Serious gamers with Quad-SLI are going to want to use SFR (Split Frame Rendering) which cuts latency quite a bit but takes a performance hit to the FPS. There are definite inefficiencies to 4-way SFR with having all four cards render portions of the same scene vs 4-way AFR. You generate 4 times as much display list info (GPU fifo data) and you have to replicate more data and uploads (if textures don't fit into the 512MB memory) across the GPUs.
I'm not sure if the Quad-SLI supports an AFR/SFR hybrid where you can have 2X2 (2 GPUs working on SFR each in AFR queue) - this might balance the performance vs latency issue better.
If you have your XP home CD, you should be able to install it.
:(
Bzzzt... most computers shipping XP home don't come with the CD - only a restore to original image program taking up a hidden partition on the HD. Think Dell, Gateway, HP etc. These users don't have backup and can't install it.
Windows Backup is not installed on the Home Edition of XP that ships by default on 90% of users' new computers. Plus if you order from Dell or somewhere nice like that, they don't give you an installation CD to install the program (which on the MS website they tell you to hunt down Ntbackup.msi on the CD and run it to install). Plus it does back-up's by folder (which means knowing where the obscure stuff is if it's not in your My documents). Plus it doesn't do selective registry backup and restore... etc.
In otherwords, it's good for saving your "My Documents" folder after you've bothered to install it.
No new features?
8 7C8}\Microsoft\Outlook Express
How about Time Machine? This is a very user friendly backup concept. Imagine a normal user performing backups on Windows. How hard is it to save off apps that rewrite 1,000 entries in your registry? Heck your average user has no idea where to find and backup his e-mail:
If you're using Outlook, they're in: C:\Documents and Settings\username\Local Settings\Application Data\Microsoft\Outlook
If you're using Outlook Express, they're in: C:\Documents and Settings\username\Local Settings\Application Data\Identities\{45E80F8D-E0D8-48D0-B459-408C2E2F
That is unless you've modified (either on purpose or accidentally) some obscure registry keys that redirect these files.
Grammar/spelling nazi mode on On a side note, you don't need to use apostrophes with acronyms when showing the plural form and the market for games on Macs is nonexistent, not inexistant. Carry on. Grammar/spelling nazi mode off.
Mr. Grammar Nazi, you are wrong on both counts. My english is far from perfect but when you wish to correct someone's grammar, it helps if you are writing proper English yourself. Both the items you attempt to correct in my post are proper English (with a capital 'E').
I direct you to the following dictionary entry from Webster's Revised Unabridged Dictionary (1913) : http://dict.die.net/inexistent/
Inexistent \In`ex*ist"ent\, a. [Pref. in- in + existent: cf. F. inexistant.] Not having being; not existing.
And it is perfectly OK to use an apostrophe with an acronym to separate the plural form. As a matter of fact, it is MORE proper to do so. However, some people have an aversion to the proper usage of apostrophes with acronyms due to the fact that it violates the rule that apostrophes are not used before an S indicating a plural. Alas, proper English has many exceptions to the rule and here is one of them, esoteric as it may be. For example, why doesn't "i before e except after c" apply to weigh or neighbor? Anyhow, I direct you to the following site at Washington State University: http://www.wsu.edu/~brians/errors/acronyms.html
God I hate being right all the time. (Just Kidding!)
I look forward to your apology and if I ever should meet you in the real world, I will expect you to buy me a beer. In the meantime, I am glad to have enlightened a Grammar Nazi's understanding of English so that he may refrain from grammar-bashing poor fellows writing proper English in his future.
I like programming video games and I like working on games for PC's and Consoles which is what I'm paid to do. I also said I think Macs are nice sexy stable platforms. Heck I even aggressively recycle and drive a fuel-efficient car too which you seem to think makes you special. However, I honestly believe that Macs are not a big enough market for $20 million dollar games to be developed specifically for them and for those games to be profitable. Game development is not the same as solving the world's hunger problems. If Mac owners are hungry for games but their market isn't profitable, it's not a crime against humanity to let them "starve" for new games. We develop games to make money, not as charity.
... in history" just shows how out of touch with reality you are. I'm perfectly happy where I am and I see no justification to risk my company's financial health with an issue that no one I work with believes in including myself. I want my games to be played by millions of people everywhere (the last one I worked on had 2+ million copies sold).
I have no personal desire to throw away my career or my company's money on developing a project projected to lose millions of dollars regardless of how sexy it is. It's hard enough to make money on PC's and Consoles as it is without targeting a niche market without a distribution channel. And equating my reluctance to push for more game development on an inexistant market to the "worst atrocities, of all kinds
There's no reason to throw away a happy life, a career, and millions of dollars on some eco-freak hippie whiners like you who think I should "grow a pair" and service them.
Macs have, depending on who you ask, anything up to a 15% market share.
Well, the people I would ask actually write industry reports for video game distribution. I'm assuming the people you ask are other Mac owners living in some fantasy world. Mac video game sales make up significantly less than 1% of the total sales of console and PC games. If they made up 15% of the total video game sales market, they would be more than PC games which make up only 12-14% of the video game market. Most game sales are on consoles (including handhelds) and then on PCs. The Macs are way out on the long end of the tail next to Linux for game sales. If Macs made as big a market share as you'd like to believe, it'd be easier to find games for the Mac than for the GameCube.
I want us to live a more sustainable lifestyle and I practice that, including recycling far more than most people do and driving a small fuel-efficient car and minimizing my driving and living close to where I work and working for a responsible company (a university). I actively encourage those I know to do the same.
There's a pretty big difference between you driving a fuel efficient car and solving the entire country's oil addiction. But dude, "grow a pair" and fix the country's energy problems by yourself. And while you're at it, "grow a pair" and bring peace to the middle east. I know you want to do it but the only thing holding you back is lack of balls.
Grow a pair and fix it.
Now there's a valid solution to all the real world problems with the Mac video game development issues. Why didn't I think of that before?
Attitudes of Mac owners like this (I'll assume you own one by your comment) do not endear you to game developers with much larger gaming markets (consoles ~ $8 billion, PCs ~ $1 billion, Macs ~ $20 million, Linux ~ $1 million) to focus on. When consoles and PC's have 99% of game sales, there's not much point to supporting other platforms -- at least here in the real world that makes sense.
Is OpenGL+SDL that bad?
When you're writing stuff for PC and XBOX (or XBOX 360), you use DirectX not OpenGL. PS2 usually uses a custom renderer or a package like Renderware. Only PS3 uses a graphics API similar to OpenGL and most people are going to bypass that for the low-level graphics API for performance reasons. Game Development IDES like Visual Studio (XBOX/XBXO360), SN (PS2/PS3), and CodeWarrior (PS2/PSP/etc) have licensing controls built-in that *ONLY* allow them to run on specific Windows PCs (either by MAC # or with floating licenses off an attached network server). Additional debugging tools (PIX, CATS, WinPACon, etc.) only run on Windows. At one time, Sony actually made a lot of their PS2 tools run on Linux but since developers had to use Windows for everything else, this effort hindered the tool usage until they made cygwin ports for windows. Finally, no one in the industry has made any serious effort to get the game-programming specific development tools we use to run on Macs.
Macs are sexy and nice, they're just not practical for my job. Sorry...
You ever think that you're in a position to, y'know, change that?
To be honest, in my current position as a game programmer for a major company, I have almost no control over the actual platforms we target for game design for current gen or next gen. These decisions are made much higher by management and marketing based on potential market and profitability. When it costs $10-20 million to make a game, no one wants to target a small unproven market. The best they'll risk is low-cost ports after the main game is written. In other words, incremental work that may increase the ROI from the already complete game. Find me a single publisher who will fund a 20-30 person team on a Mac game for two or more years because a programmer told them to do so and I'll eat my words. That's the effort going into high-end PC and console games right now and the sad facts of the economics.
I desire my PC to be pure of spyware, security flaws and unstability.
Somehow Macs do this and manage to be sexy at the same time. That said, I have a PC at work and home. I'd buy a Mac for home except that Macs are sadly lacking in the one field in which I work (video game development).
4) At this point, you have no debt, and you have reached your "margin of safety" amount. Once you have built up an additional $3k to $5k on top of your margin of safety, open up a discount brokerage account (e.g., E-Trade).
Don't use E-Trade. They have so many fees it's impossible to count them all. They charge $19.95 per trade plus some other fees (ends up about $23-25 a trade for a casual investor), they charge to get money out, and they charge you about $60 to CLOSE your account!!! Try Scottrade (www.scottrade.com). Scottrade has lots of local offices (three of them within 5 miles of my house in the NW burbs of Chicago) and they only charge $7 a trade. Plus Scottrade has better margin rates and will pay higher interest (although still low) on any cash sitting in your account.
FWIW, I'm a PS3 developer and the preferred method (i.e. generates the fastest code for the general case) for programming SPU's is to use C with intrinsics. C mind you... not C++. Heavy use of esoteric C++ features on the SPU's just generates some of the worst performing binary code you might ever see. Plus, there are lots of compiler specific (GCC) tricks though. A great place to see some of the performance tricks on CELL is on PS3 developer Mike Acton's site http://www.cellperformance.com/
I prefer to use general optimizations first over compiler specific ones but if top performance is required (and in video games it often is) you have to occasionally make sacrifices of portability and code readability for speed in low level routines.
US Secretary of Education Rod Paige, a member of the Bush Administration, called the National Education Association, the nation's largest teachers union, a terrorist organization. AFAIK, they've never committed hostage-taking, bombings, death threats, or other terrorist activity. Their biggest crime is using strike tactics for collective bargaining and closing schools when to trying to get better pay and working conditions for our nations Teachers.
Around 300 AD, Christian riots and violence were widespread throughout parts of the Roman Empire to a degree that was disrupting society on a regular basis. Think Paris-car burnings combined with Middle East Anti-American protests on a large scale. Now imagine that happening in the US and tell me what the Bush Administration would call it.
So what do you call the Inquisition and the Crusades? Christianity hardly spread throughout the world by peace. Even in Ancient Rome 2000 years ago, the early Christians caused enough destruction and disturbances that the Bush administration would call them Terrorists today.
what does science have to do with history?
The Bible is a historic document in the same way that the Iliad is a historic document. Both are collections of myths and fables that are roughly based on actually occurences in history. Both have supernatural events that were not likely but make for a better story. Science lets us determine which parts are likely to be true (i.e. history) and which parts are likely to be nothing more than myth.
Yes, the likelihood of a group of Jewish fisherman making up a story about a Messiah figure who claimed to be God (blasphemy) and then turning the entire Roman empire upside down in the matter of a few decades is highly unlikely. It is even more unlikely that they would all suffer torture and death to protect a story that is not true. And yet, that is exactly what happened. If anything, this is a strong indication that their story was real. Would you die for something you know to be false?
By your logic, a prophet in the middle east who turned the entire regious upside down, resulting in the rapid conversion of the entire area to the same belief must be correct. ESPECIALLY since he has thousands of men and women lining up to die for his beliefs on a daily basis and receive martyrdom for their cause.
Yes... there you have it. Following your logic, both Christianity and Islam are true. And since Muhammed came afterward Jesus and plenty more people are willing to die for Muhammed, Islam must be "more true" than Christianity.
Do you see the flaws in your logic now or are you converting to Islam?
When people require absolute faith regardless of the overwhelming contrary evidence, they have already sacrificed enough of their own identity and ability to reason that sacrificing their lives is merely the next step of losing themselves to their beliefs. Welcome to the Church of Jim Jones, you'll enjoy the Kool-Aid.
"[...]a few people might complain about losing San Francisco and New York to rising water levels."
As a species though, even as a culture, the long term (100+ years perspective) impact would be minimal. We'd rebuild inland and carry on.
The problem with non-linear systems like the climate is that it's possible that we'd get something like a "Venus-effect" eventually and that might be a bit more threatening than displacing a few hundred million people with flooding.
The fast C code that's executing has already been written.. some of it is in the python interpreter, some it is in the ksh and php interpreters, most of it is in the db2 interpreter.
Not *ALL* the "fast C" code has been written. There's lots of code out there that needs to run faster... some written and some yet to be written. You can use python to string together all the calls to already-written-C-libraries you want but when you have an inner loop of a tight algorithm that isn't in a library fo you, you're going to have to break down and code it in C (or JAVA running through a JIT-native compiler, etc).
Go to Walmart and pick a random item off the shelf. You have about a 70% chance it was made in similar conditions in China.
It's not broken - it's that way by design and there's nothing to "fix" in order to make a good game on the PS3. There's nothing to "fix" because even if you need to share large buffers between the SPU's and RSX, there's a simple and quite fast copyback mechanism.
The SPU single-read from RSX memory is only meant to be used rarely for occasional one-time reads. What Sony did in alerting developers to the faster method of dealing with larger shared buffers is the right thing to do. Not to mentionat that since a fast read out of video memory is not a feature that other consoles have or even most PC cards support, any algorithms making extensive use of this feature are probably being written from scratch anyway and can very easily incoporate the fast RSX copyback. This is analogous to DirectX on a PC, where when you "lock" a surface for reading, the driver will often do the equivalent copyback from video memory to system memory without you even knowing about it! On the PC, it may even de-swizzle or change formats during the copyback.
Pay attention. The article says that SONY is telling the developers to avoid using local memmory at all - that means, it won't be fixed in the retail version.
There are three types of directly addressable memory on the PS3 and there are three types of processors.
PROCESSOR == MEMORY
PPU == main memory
SPU == (SPU) local memory
RSX == (RSX) local memory (AKA video memory)
What Sony is telling developers NOT to do is to READ from (RSX) local memory (AKA video memory) using the SPU. This would be much more clear if the article mentioned this was local memory on the RSX not the SPU. The PS3 has a very fast write port from SPU to RSX. There is no fast read port from RSX to SPU, and for the most part it isn't necessary.
Most of the time, developers will be using the SPU to generate large data sets (dynamic geometry, etc) to be fed to the RSX similar to how the VU1 fed data to the GS on PS2. For the rare times that the SPU will need to access RSX, the main memory writeback mechanism will suffice for large data accesses (i.e. post process on an entire bitmap) and the slow path can be used for very occasional, single small reads (i.e. check the color of just one pixel). On the PS2, there isn't even a slow path for the reads from video memory to VU1. You'd have to set up a copy-back from GS-memory to main memory using the EE and the write it back into VU1 memory (again using the EE or setting up a DMA with the EE) with lots of waits or interrupt driven synchronizations.
FWIW, I don't believe the XBOX360 has a fast ability for the PPC's to read from their local video memory (embedded on the ATI GPU) either.